EP3449512A1

EP3449512A1 - Method for producing rear surface contact solar cells from crystalline silicon

Info

Publication number: EP3449512A1
Application number: EP17716901.8A
Authority: EP
Inventors: Morris DAHLINGER; Kai Carstens; Juergen H. Werner; Juergen Koehler; Sebastian Eisele; Tobias Roeder; Erik Hoffmann
Original assignee: Universitaet Stuttgart
Current assignee: EnBW Energie Baden Wuerttemberg AG
Priority date: 2016-04-27
Filing date: 2017-04-12
Publication date: 2019-03-06
Anticipated expiration: 2037-04-12
Also published as: DE102016107802A1; CN109314151A; US20190348560A1; JP2019515498A; EP3449512B1; WO2017186488A1

Abstract

A rear contact solar cell (10) comprises a wafer (16) with an antireflection coating (12) on the front face, an emitter (20) and a back surface field (32) on the rear face, and contacts (28), created by laser ablation, on the rear face, the maximum pitch being 800 micrometers. The invention further relates to a method for producing a solar cell of said type.

Description

Process for the production of reverse-contact solar cells made of crystalline silicon

The invention relates to a method for producing back-contacted solar cells made of crystalline silicon.

Known manufacturing processes use furnace diffusion for the production of n-type doped and p-type doped regions and for contacting vapor-deposited metals. Masking steps are necessary for the structured production of the doped regions required for back-contacted solar cells as well as for the metallization. Since the silicon wafer in the diffusion furnace has the same temperature everywhere, the diffusion takes place uniformly over the entire surface. Generation of differently doped stripe or dot structures of the n- and p-type regions on the back side of the solar cell therefore requires, for each of the diffusions, either masking, which locally prevents diffusion of the doping atoms, or else a local etching step after diffusion. to remove the area not to be diffused. In both cases, both the application of a diffusion-inhibiting masking layer or a etch-resistant protective layer as well as their high-resolution structuring required. Since both boron diffusion and phosphorus diffusion must occur locally, these steps are necessary prior to performing the furnace diffusion and must additionally be aligned with high precision. Also, an opening of a backside passivation layer for contacting the solar cell requires high precision, so that a lithography step is required. Furthermore, the application of the metal contacts additionally requires at least one lithography step. If two different metals are used, two lithography steps are necessary.

For the reasons presented above, the production is back contact

Solar cells using masking technique on lithography not economical.

According to WO 2007/081510 A2, a back-contacted solar cell made of crystalline silicon is prepared by printing precursor layers for subsequent furnace diffusion locally by means of screen printing or inkjet printing.

Such a production leads to inaccurate tuning of the doped regions and thus to a non-optimal efficiency.

It is basically known from DE 10 2004 036 220 A1 to generate regions doped by laser doping on solids with a high degree of freedom from defects. In this case, a medium containing a dopant is first brought into contact with a surface of the solid. Subsequently, by irradiation with laser pulses, a region of the solid beneath the surface contacted with the medium is briefly melted, so that the dopant diffuses into the melted region and recrystallizes the melted region without defect during the cooling process.

In principle, masking steps and lithography steps for doping by furnace diffusion can be avoided with such a method. There remains the problem of a simple and inexpensive contacting at the back of the solar cell. From WO 2015/071217 A1 it is known to use a laser doping method for the production of the doped regions. By laser ablation contact surfaces are exposed at the back of the solar cell, which are then contacted by screen printing (see also M. Dahlinger, et al., "Laser Doped Back-Contact Solar Cells", IEEE Journal of Photovoltaics, Vol. No. 3, May 2015, pp. 812-818, and M. Dahlinger, et al., "Laser Doped Screen-printed Back Contact Solar Cells Exceeding 21% Efficiency", Energy Procedia, Vol. 55, September 2014, p. 410-415).

Screen printing is a cost-effective and well-tried one

Method, however, can not be achieved by screen printing high precision, whereby the efficiency is limited. Furthermore, the recombination of charge carriers in screen printing is higher than, for example, at approximately PVD (physical vapor deposition) deposited contacts.

From WO 2006/042698 A1 it is known to first produce a metal layer on the rear side for contacting a backside-contacted solar cell, then to deposit an etching barrier layer, then selectively remove it by means of a laser and finally by an etching step the electrical separation between to reach the different polarities.

Basically, however, a further increased efficiency is sought.

According to WO 2015/047952 A1, a metal foil is applied for contacting. The film is selectively welded by a laser and separated between the different polarities.

Such a method is very time consuming and costly.

According to P. Verlinden et al., "High Efficiency Large Area Back Contact Concentrator Solar Cells with a Multilevel Interconnection", International Journal of Solar Energy, 1988, Vol. 6, pp. 347-366, it is known to use back-contacted solar cells using anodization process to make the contact. The process is very complicated because of various steps including various lithography steps.

Finally, it is known from US 2016/0020343 A1 to transmit dopants or electrically conductive material by means of a laser transfer process, for example to produce a finger structure for contacting.

Transfer by laser transfer is a rather expensive process.

In addition, a further increased efficiency is sought. Furthermore, a laser transfer is usually limited to seed layers, ie thin layers of a few 10 nanometers thickness. These are not sufficient as a metallization for a current transport and usually have to be thickened subsequently, which requires a further process step.

Against this background, the invention has the object, a method for

Creating back-contacted solar cells made of crystalline silicon, which allows the simplest possible and cost-effective production with high quality and the highest possible efficiency.

This object is achieved by a process for the production back contact

Crystalline solar cells solved with the following steps:

(a) doping to produce an n-type or p-type doped region, preferably by laser doping;

(b) exposure of contact surfaces on the back of the solar cell, preferably by means of laser ablation;

(c) applying a metal layer to a back side of the solar cell; and

(d) patterning the metal layer by laser ablation to create metallic contacts, the pitch being at most 800 microns. The object of the invention is completely solved in this way.

Since the production of contacts at the back of the solar cell by a

Laser ablation step according to (d), it is possible due to the high precision to obtain a small pitch which is at most 800 microns, preferably at most 500 microns, more preferably at most 100 microns, more preferably at most 60 microns. For example, it may be a pitch of about 50 microns.

It was inventively recognized that the efficiency increases, the smaller the pitch.

In order to allow a pitch as low as possible, the other steps in the manufacture of solar cells, such as the production of doped regions and the exposure of contact surfaces, possibly avoiding lithography and masking steps and avoiding printing techniques, should be as high as possible To ensure precision. Preferably, the laser technology is used for this purpose.

Depending on the precision of the adjustment of the laser used, the pitch is limited to the bottom. A lower limit represents a pitch of about 5 microns.

It is understood that the term "solar cell" is to be understood in its broadest sense, including special forms such as photocells.

In a further embodiment of the invention, an etching-resistant layer is first applied after step (c), which is selectively removed in step (d), and wherein by a subsequent etching step, the mutually electrically isolated, metallic contacts are generated. In this way, short circuits between adjacent contacts are safely avoided without the risk of damage by too high penetration depth of the laser during ablation.

In an alternative embodiment of the method according to the invention, an aluminum layer is applied in step (c), then a layer resistant to anodization applied, which is selectively ablated in the subsequent step (d) by laser and then completely anodized in the ablated areas.

In this way, instead of complete removal of the aluminum present between adjacent polarities, complete conversion to alumina is achieved, which also safely excludes short circuits.

A certain advantage results from the remaining flat surface.

In an additional embodiment of the invention, the metallic contacts are interconnected by busbars, which consist of strips of metal foil, which are contacted with the interposition of at least one dielectric layer by means of laser welding through the dielectric layer.

According to a first embodiment of the invention strips are made of anodized

Aluminum foil used to create the busbars. The laser welding process takes place through the insulating layer to one polarity each.

Of course, another dielectric layer or stack may also be used on the foil strip or on the backside of the wafer for insulation.

To generate a p-type emitter and / or to generate an n-type back surface field (BSF) on the back side of the solar cell, a laser doping step is preferably used. For this purpose, preferably first a precursor layer containing a dopant, in particular boron, aluminum or gallium deposited on the back of the solar cell and a p-type emitter produced by local irradiation by means of a pulsed laser.

Alternatively, the p-type emitter may be generated locally by ion implantation with a dopant, particularly boron, aluminum or gallium.

Also in this way, the emitter can be produced without masking or lithography steps with high precision.

According to a further embodiment of the invention, a higher doping is locally generated under the emitter contact surfaces in the emitter doping by means of laser irradiation by beam shaping or by using a further independently focused laser beam. In the case of beam shaping, it is crucial that the pulse energy density in the region of the contacts is locally increased in order to obtain a higher doping there. A corresponding beam shaping can, for. B. by means of a diffractive optical element.

In this way, a particularly low-loss contact is made possible in a simple manner.

For laser doping, a pulsed laser is preferably used, preferably with a pulse duration of 30 nanoseconds to 500 nanoseconds, more preferably with a wavelength of 500 to 600 nanometers, more preferably with a pulse repetition rate of 1 kHz to 2 MHz, further preferably with a pulse energy density of 1 J / cm ² to 5 J / cm ² .

The use of such a laser results in optimum matching to the doping task. In this way, the silicon surface and the precursor layer can be locally heated to the extent that the doping process can be carried out locally to the desired depth in the shortest possible time, at the same time avoiding overdoping. By a local variation of the pulse energy density, the Doping optimally adapted both in the contact areas, as well as in the non-contacted areas of the emitter.

Preferably, the laser beam is focused on a rectangular area X by means of an optical system.

Y and the laser and substrate are incrementally moved relative to each other by one step length L to dope predetermined areas.

In this way, a precise doping in rectangular or linear

Areas are generated.

In this case, the width X is preferably 0.02 to 2 millimeters, while the length Y is preferably between 5 microns and 500 microns.

Preferably, the stride length L by which the substrate and the laser are incrementally moved is between 0.1 · Y and Y. By repetitively irradiating and displacing the silicon wafer or by displacing the laser beam imaged on the surface in Y. Direction by the stride length L, the entire desired area of a stripe or dot is doped.

In an advantageous embodiment of the invention is for generating an n-type back surface

Fields (BSF) on the back of the solar cell, first a phosphorus silicate glass layer (PSG) deposited as a precursor on the substrate, which is then irradiated by means of a laser to produce an n-type doping. The deposition of the PSG layer occurs simultaneously with a front surface field (FSF) of the wafer in a high temperature diffusion furnace. The generation of an FSF allows for improved passivation of the solar cell front.

Preferably, the phosphorus silicate glass layer is removed after the laser doping by etching and then partially etched back the phosphorus doped layer at least on the back of the substrate. The etching back is done, depending on the depth and phosphorus concentration on both sides of the silicon wafer or only on the back. The purpose of the etchback step is to reduce the phosphorus present in the boron emitter regions. The phosphorus surface concentration in the emitter region can be adjusted by the re-etching step so that it is at least fivefold smaller than the boron surface concentration after a subsequent thermal oxidation.

A reduction of the phosphorus concentration on the front is required, if this is too high phosphorus doped. The aim here is a phosphorus surface concentration of about 1 × 10 ¹⁸ cm ^-3 to 1 × 10 ²⁰ cm ^-3 after a subsequent thermal oxidation step for optimum front side passivation by the FSF thus produced. In addition, chemical etching of the silicon wafer is achieved by the etching back.

After the laser doping of the BSF layer or after the partial back-etching step, a thermal oxidation in the range of 700 ° C to 1 100 ° C, preferably 800 ° C to 1050 ° C, performed.

In this so-called drive-in step, silicon dioxide grows as surface passivation.

Furthermore, due to the high temperatures, the doping atoms continue to diffuse into the silicon wafer. As a result, the surface concentration of the doping decreases both in the solar cell emitter and in the BSF and FSF.

In a further preferred embodiment of the invention, an anti-reflection layer is deposited on the front, preferably a silicon nitride layer deposited by means of PECVD.

On the back side of the solar cell, a stacked layer of low-silicon and silicon-rich silicon oxide or silicon nitride is preferably deposited by means of PECVD. The low-silicon layer preferably has a low refractive index (n <1, 7) and a thickness between 70 nanometers and 300 nanometers, while the following silicon-rich layer preferably has a layer with a high refractive index (n> 2.7) and a thickness between 10 nanometers and 100 nanometers. Both layers can be deposited one after the other in the same process step in the same plant. They increase, among other things, the "light trapping" and passivate the back. Furthermore, the high refractive index layer serves as ablation masking step in the subsequent process steps.

After the application of the stack layer is preferably by means of a UV laser a

Ablation performed to expose the areas to be contacted, wherein preferably only the last deposited silicon-rich layer is ablated in the areas to be contacted, since only this absorbs the UV radiation. The low-silicon layer is transparent to UV radiation and therefore can not be absorbed by it, thereby preventing its ablation.

The remaining layer up to the silicon interface can then for a subsequent

Contacting be etched away.

In this way, there is a local opening of the contact surfaces without a

Laser damage to the silicon surface.

On the front side of the solar cell is preferably before the doping of the emitter

Front side texture generated. This can be done by wet-chemical polishing and texture etching of the substrate at the front.

The wet-chemical polishing can be used here as a first step, if necessary also on one side,

which is followed by a one-sided wet-chemical texture etching. The sequence can also be reversed by first starting with a wet chemical texture etch to create the front side texture of the solar cell, followed by a wet chemical one-sided polishing of the back side of the solar cell and depositing a boron-containing precursor layer on the back side of the solar cell. [0059] A back-contacted crystalline silicon solar cell fabricated by the above-described method comprises a wafer having an anti-reflection layer on the front side, an emitter and a back surface field on the back, and laser ablation Contacts on the back, with a maximum pitch of 800 microns. Preferably, the pitch is significantly less, such as in the range of 100 microns or less, such as about 50 microns.

This results in a high efficiency. In addition, the dependence on the BSF proportion / _B SF is reduced.

It is understood that the above and the still to follow

explanatory features of the invention not only in the particular combination, but also in other combinations or alone, without departing from the scope of the invention.

Further features and advantages of the invention will become apparent from the following description of a preferred embodiment with reference to the drawings. Show it:

1 shows a simplified cross section through a solar cell according to the invention;

FIG. 2 shows a schematic representation of the top view of a unit cell of the back side of the solar cell according to FIG. 1; FIG.

3a-f the efficiency of a solar cell as a function of the BSF content / BSF and the pitch p of the solar cell for different wafer qualities and

Fig. 4 is a schematic representation of the connection of contacts by means

Foil strips over welds. In Fig. 1, the cross section of a solar cell according to the invention is shown schematically and generally designated by the numeral 10.

The solar cell 10 has an n-type silicon wafer 16. At the front of this is provided with a passivation and anti-reflection layer 12 on a pyramid-like texture. Below this is a front-side phosphorus diffusion layer, the front surface field (FSF) 14.

On the rear side, the solar cell 10 has laser-doped boron emitter regions 20, on each of which selectively more heavily doped emitters 18 are formed, on which contacts 28 are applied.

On the back of the solar cell 10 are also laser doped by means of phosphorus

Base regions 22. The backside is insulated by a passivation layer 24 opposite the contacts 28, by which the contacting with the selectively doped emitters 18 and the heavily doped base regions 22 is established.

Fig. 2 shows a schematic representation of the top view of a unit cell of the back of the solar cell 10 of FIG. 1. At the top and bottom of the drawing, the unit cell is mirrored continued. Here is the left and right of the solar cell edge. 30 shows the basic contact area. 22 indicates the base region created by the BSF doping (Back Surface Field). 34 denotes the doping for the current busbar (bus bar). 20 denotes the emitter doping. 18 denotes the selectively higher doped emitter. Finally, 36 designates the emitter contact region.

The so-called "pitch" denotes the distance between two adjacent emitters 18 (the

Pitch is, so to speak, the "period" of the solar cell.) In Fig. 1, the pitch is designated by p.

FIGS. 3a-f show the dependence of the relative efficiency of a solar cell in FIG

Dependence on the pitch p and on the BSF fraction / _B SF (area fraction of the ohmic contact (base region 22, BSF doping) relative to the total area (base region 22 plus emitter 18)), for different wafers. Where p is the resistivity of the wafer and τ is the volume life of the minority carriers.

It can be seen that irrespective of the specific resistance of the wafer and of the volume lifetime, the smaller the pitch p, the greater the relative efficiency η. In addition, the smaller the pitch p, the lower the dependence on the BSF component / _B SF. The maxima pronounced at a larger pitch p as a function of the BSF component / BSF are smoothed with a smaller pitch p. At a pitch p of 50 microns there is practically no longer any dependence on the BSF content / BSF

This dependence on the pitch p is utilized according to the invention in order to achieve the highest possible efficiency.

With the aid of laser technology, it is possible according to the invention to produce a solar cell 10 with a pitch <800 micrometers, preferably <100 micrometers, more preferably <60 micrometers, in a technically relatively simple manner. In general, the pitch is greater than 5 microns.

The manufacture of such a solar cell 10 will be described in detail below.

The inventive method is completely without masking.

Instead, laser doping steps and a laser ablation step are used to open the backside passivation layer. Optionally, the laser doping to create the emitter may also be replaced by a local ion implantation step. Another laser ablation step is used in making the contacts 28 for emitter 18 and base regions 22.

To produce the solar cell 10, an already ground-doped n-type silicon wafer is used.

On the front side of the solar cell 10, a wet-chemical alkaline texturing is first carried out to produce a pyramid-like textured surface. Subsequently the back of the solar cell 10 is wet-chemically polished on one side (alkaline or acidic). This is followed by the deposition of a boron, aluminum or gallium-containing precursor layer on the rear side of the solar cell 10. The sequence of these steps can also be reversed: First, a wet-chemical polishing (possibly also one-sided) can be carried out, followed by a one-sided wet-chemical texturing on the front side of the solar cell 10.

The precursor layer on the back side of the solar cell 10 may be e.g. with the aid of a sputtering system, or a plasma-chemical precipitator, e.g. APCVD, or by means of a spin-coating method or a spray-coating system.

Subsequently, a p-type emitter is generated on the back of the solar cell 10 by means of a laser doping process. In this case, a laser pulse melts the surface of the silicon wafer. Due to the high diffusion constants in liquid silicon, the doping atoms present in the precursor layer diffuse into the surface of the silicon wafer during the liquid phase within about 100 nanoseconds to a depth of about 1000 nanometers, thus forming the p-type emitter.

In this case, the laser beam is imaged on the silicon surface with the aid of optics such that a single laser pulse melts a sharply delimited rectangular area with an area of the size X.Y. Preferably, 0.02 millimeter <X <2 millimeter and 5 micrometer <Y <500 micrometer. Here, the size X defines the width of the emitter strips or dots. By repeatedly irradiating and displacing the silicon wafer or by moving the laser beam imaged onto the surface in the Y direction by the step length L, the entire surface of an emitter stripe or dot is doped. In this case, preferably, 0.1 · Y <L <Y.

In addition, emitter doping produces locally increased boron doping below the emitter bus bar region. This is done either by beam shaping during laser irradiation or by using a further, independently focused laser beam. When shaping the beam, it is crucial that the pulse energy density in the Area of the contacts is increased locally, in order to obtain a higher doping there. A corresponding beam shaping can, for. B. by means of a diffractive optical element.

Due to the locally increased boron doping below the emitter bus bar region (also called selective emitter), a reduced overall series resistance and thus a better filling factor of the solar cell is achieved. The locally increased boron doping below the emitter contact further reduces the recombination of charge carriers at the metal-semiconductor interface. This increases the open circuit voltage and thus the efficiency of the solar cell 10. Further, the contact resistance is reduced, whereby the total series resistance decreases and the fill factor increases.

Both local dopants can be generated without additional process step during emitter laser doping. By varying the laser pulse energy density, the doping profile and thus the sheet resistance are adjusted.

After the laser doping of the emitter 18, the remaining precursor layer is removed wet-chemically. The chemical solution used depends on the precursor layer used.

Subsequently, the silicon wafer 16 is cleaned by a hydrochloric acid-hydrogen peroxide solution and then in a hydrofluoric acid bath.

As an alternative to the above-described laser doping using a previously written off precursor layer, the boron-doped emitter 18 of the solar cell 10 can also be produced by means of a local ion implantation step. A defect-free recrystallization of the amorphized by the ion implantation of silicon and the activation of the doping atoms is achieved by the thermal oxidation later described, which also follows in the case of a laser doping step. On the back side of the silicon wafer is also a so-called. Back Surface Field (BSF) in

Form of a highly doped n-type region produced by laser doping using a phosphorus-rich precursor layer.

For this purpose, first in a standard tube high temperature furnace, a phosphor rich

Phosphorus silicate glass layer deposited on both the front and on the back of the silicon wafer. Here POCI ₃ and 0 ₂ serve as process gases. The deposition takes place at temperatures between 700 ° C and 850 ° C. In addition, a part of the phosphor diffuses a few tens of nanometers to 500 nanometers into the silicon wafer. In this case, the diffusion is optimized in such a way that a doping which is as shallow and low as possible takes place, but nevertheless a phosphorus-rich phosphosilicate glass is formed or a phosphorus-rich interface is present.

The phosphorus-rich interface or the phosphorus-silicate glass layer serves as a doping source for a subsequent laser doping process.

As described above for the emitter doping, a laser pulse here melts the

Surface of the silicon wafer on. Due to the high diffusion constant in the liquid silicon, the phosphorus atoms present in the phosphorus silicate glass layer diffuse into the surface of the silicon wafer during the liquid phase within about 100 nanoseconds to a depth of about 1000 nanometers and form the BSF region 32 , a highly doped n-type region. As described above, in this case the laser beam is imaged on the silicon surface with the aid of optics such that a single laser pulse melts a sharply delimited rectangular area with an area of the size X.Y. Again, by a relative movement between silicon wafer and laser beam in the Y direction by the stride length L gradually the entire surface of a BSF stripe or point is doped. Incidentally, the same geometric conditions are used here as described above in connection with the emitter doping.

The phosphosilicate glass layer becomes after local BSF laser doping

Hydrofluoric acid solution (1% to 50%). Subsequently, the phosphorus doped layer at least on the back of the

Substrates partially etched back. For this purpose, a wet-chemical solution of hydrofluoric acid, nitric acid, acetic acid and deionized water is used to etch back about 10 nanometers to 300 nanometers of the phosphorus-doped layer at depth. This etching step takes place depending on the depth and phosphorus concentration on both sides of the silicon wafer or only on the backside. The purpose of the etchback step is to reduce the phosphorus present in the boron emitter regions.

The emitter surface phosphorus concentration should be at least fivefold less than the boron surface concentration after the thermal oxidation described below. The reduction of the phosphorus concentration on the front side is required if it is too high phosphorus doped. The aim here is to obtain a phosphorus surface concentration of 1 × 10 ¹⁸ cm ^-3 to 1 × 10 ²⁰ cm ^-3 after the following high-temperature oxidation. In addition, the back etching step serves for the chemical cleaning of the silicon wafer.

Subsequently, a wet-chemical cleaning is first carried out by a hydrochloric acid-hydrogen peroxide solution with a subsequent hydrofluoric acid.

This is followed by thermal oxidation as a so-called drive-in step. Here, a silicon dioxide layer grows as surface passivation. Alternatively, a silicon nitride layer, a silicon oxynitride layer or a silicon carbide layer stack may also be used. During drive-in, the doping atoms continue to diffuse into the silicon wafer due to the high temperatures (about 800 ° C to 1050 ° C). As a result, the surface concentration of the doping decreases both in the back surface field (BSF) (base area) and front surface field (FSF), as well as in the emitter. The resulting silicon dioxide grows up to a layer thickness of 5 nanometers to 105 nanometers, which in combination with another anti-reflection coating layer thicknesses in the range of 5 nanometers to 20 nanometers are aimed.

In order to reduce the effective reflection of the solar radiation on the surface of the solar cell 10, a silicon nitride layer is formed on the front side of the solar cell 10 plasma enhanced vapor deposition (PECVD). The refractive index should be between 1, 9 and 2.3.

On the backside of the solar cell 10, a 1 to 50 μm thick layer of aluminum, e.g. applied by evaporation or sputtering. This layer serves for later generation of the contacts 28 on the base regions 22 and the emitters 18.

On the aluminum layer is a metallic, semiconducting or dielectric

Cover layer applied by about by evaporation, APCVD, PECVD, CVD or cathode dusts. This layer should be etch resistant or only slightly etchable to a subsequently used etchant (such as phosphoric acid, hydrochloric acid, sodium hydroxide or potassium hydroxide). It can e.g. nickel, zinc, amorphous silicon or SiOx, silicon nitride or silicon carbide.

Now there is a local laser ablation of the applied to the aluminum layer cover layer.

In a subsequent etching step, the aluminum in the laser-exposed areas is removed by means of an etchant (for example phosphoric acid, hydrochloric acid, sodium hydroxide or potassium hydroxide), so that isolated contacts 28 result on the base regions 22 and the emitters 18.

Instead of producing an insulation between the alternating contact areas by removal by means of etching, according to a variant of the method, an insulation can be produced by selective anodization of an aluminum layer.

For this purpose, after the application of the aluminum layer, a layer resistant to an anodization, for example SiO _x , SiN _x , SiC _x , Si, Ni, Cu, is applied. This is selectively removed in a subsequent step by means of laser ablation. Subsequently, the ablated areas are treated in an anodizing bath (for example H ₂ S0 ₄ or oxalic acid). Re) completely anodized (in Fig. 1, the existing between adjacent contacts 28 slots would be completely filled with alumina in this variant).

In any case, a very small pitch p is made possible by the use of laser technology, which may be on the order of 100 μηη or even in the range of about 50 μηη. As can be seen from FIGS. 3a-f, the efficiency η is thereby significantly improved.

In the laser doping processes described above, a pulsed laser system is used (see WO 2015/071217 A1 and DE 10 2004 036 220 A1, which are incorporated herein by reference in their entirety). For generating an optimized depth profile of the dopants, the following laser parameters are preferred:

Pulse duration between 30 nanoseconds and 500 nanoseconds,

Wavelength between 500 nanometers and 600 nanometers,

- Pulse repetition rate between 1 kHz and 2 MHz,

- Pulse energy density between 1 J / cm ² and 5 J / cm ² .

Optionally, the busbars (contact tracks) 34 for the further cell interconnection are produced by laser welding of both contact polarities (emitter and base) with film strips of a metal foil. The foil strips may overlap the other polarity. A dielectric layer or a layer stack isolates the film strips from the complementary polarity.

In the simplest case strips of aluminum foil are used, which is provided on the contacts 28 side facing with an insulating anodization. The laser welding process takes place through the insulating layer to one polarity each. Of course, another dielectric layer or stack may also be used on the foil strip or on the backside of the wafer for insulation.

Claims

claims

Process for the preparation of back-contacted crystalline silicon solar cells, comprising the following steps:

(a) doping to produce an n-type or p-type doped region (20, 22), preferably by laser doping;

(B) exposure of contact surfaces (26) on the back of the solar cell (10), preferably by means of laser ablation;

(c) applying a metal layer to a back side of the solar cell; and

(d) patterning the metal layer by laser ablation to produce metallic contacts (28), the pitch being at most 800 microns.

The method of claim 1, wherein the doped region (20, 22) is generated by laser doping.

The method of claim 1, wherein the pitch (p) is at most 500 microns, more preferably at most 100 microns, more preferably at most 60 microns.

The method of claim 1 or 2, wherein the pitch (p) is at least 5 microns.

Method according to one of the preceding claims, wherein after step (c) first an etch-resistant layer is applied, which is selectively removed in step (d), and wherein by a subsequent etching step, the mutually electrically separated, metallic contacts are generated.

6. The method according to any one of claims 1 to 4, wherein in step (c) an aluminum layer is applied, then a resistant to an anodization layer is applied, which is selectively ablated in the subsequent step (d) by means of laser and then ablated in the Areas completely anodized.

7. The method according to any one of the preceding claims, wherein the metallic contacts (28) are interconnected by busbars (34) consisting of strips of metal foil, which are contacted with the interposition of at least one dielectric layer by means of laser welding through the dielectric layer therethrough.

A method according to claim 7, wherein strips of anodized aluminum foil are used to create the busbars (34).

9. The method according to any one of the preceding claims, wherein for generating a p-type emitter (20) and / or for generating an n-type back surface field (BSF) (22) on the back of the solar cell (10) a Laserdotierschritt is used.

10. The method according to any one of the preceding claims, wherein a precursor layer containing a dopant, in particular boron, aluminum or gallium, on the back of the solar cell (10) is deposited and a p-type emitter (20) by local Irradiation is generated by means of a pulsed laser.

1 1. Method according to one of claims 1 to 9, wherein a p-type emitter (20) is generated locally by ion implantation with a dopant, in particular boron, aluminum or gallium.

12. back-contacted solar cell, preferably prepared according to a

preceding claims, comprising a wafer (16) with an anti-reflection layer (12) on the front side, with an emitter region (20) and a back surface field (22) on the back, and with laser ablation. placed contacts (28) on the back, with the pitch (p) at most 800 microns.

13. A solar cell according to claim 12, wherein the pitch (p) is at most 500 microns, more preferably at most 100 microns, more preferably at most 60 microns, wherein preferably the pitch (p) is at most 5 microns.

14. Solar cell according to claim 12 or 13, wherein the contacts (28) of the base regions (22) and emitter regions (18) by bus bars (34) of metal foil strips are connected, which are electrically connected by laser welding points (38) through a dielectric layer ,

15. Solar cell according to claim 14, wherein the metal foil strips (34) consist of anodized aluminum foil.